Global Demographics & Retirement Implications - State Street ...

Page created by Gerald Fletcher
 
CONTINUE READING
Demographics & Retirement Research

Global Demographics
& Retirement Implications
Amlan Roy                   Key Points
Global Chief Retirement
Strategist
                            • Multiple demographic changes are having a big impact on how people work,
                              consume and save. This creates challenges for governments, institutions and
amlan_roy@ssga.com
                              individuals in ensuring retirement systems are sustainable

Amy Le                      • We highlight significant demographic heterogeneity across ten advanced
Investment Strategist,        countries (Australia, Canada, France, Germany, Italy, Japan, Netherlands,
Global Demographics           Sweden, the UK and the US) and show how changing age boundaries can lower
and Retirement Research       old-age dependency ratios. We also explore the role demographic differences
amy_le@ssga.com               play in explaining differences in consumption, savings, labour, retirement and
                              debt patterns across countries
Catherine Reilly            • Creating sustainable systems that reflect modern demographics requires policy
Global Head of Research,      changes, e.g., abolishing mandatory retirement ages, making state benefits
Defined Contribution
                              more flexible, promoting gender equality, adopting selective immigration and
catherine_reilly@ssga.com
                              encouraging lifelong learning

Alistair Byrne              • The shift towards Defined Contribution pensions risks increasing income
Head of European
                              inequality in retirement, so policy changes must account for the increasing
Investment Strategy,          heterogeneity of the population
Defined Contribution        • Offering retirees appropriate retirement income options and default solutions
alistair_byrne@ssga.com       should aid the transition from the savings phase to the retirement phase

                            Contents
                            1. The Evolving Demographic Landscape
                            2. Changing Age Boundaries for Dependency Ratios
                            3. Macro Implications of Demographic Changes
                            4. How Retirement Systems Can Adapt to Demographic Changes
                            5. Conclusion
                            References
                            Appendix
Demographics & Retirement Research | Global Demographics & Retirement Implications

Introduction
Global changes in demographics, behaviour and policy are dramatically
transforming the retirement landscape. In this report, we examine how
demographics, in particular, are having an impact on the growth and evolution
of retirement savings and pension systems. As articulated by Peter Drucker1 and
others, there is a growing consensus that most long-term pension promises and
entitlements are now unsustainable and that reform is urgently needed.

Our view of demographics is a broad one. It encompasses a range of “consumer
and worker characteristics” that go beyond the narrow interpretation of
demographics as being solely age-related. Evidence from various sub-disciplines
of economics shows that people of the same age can behave very differently
in terms of consumption, savings and risk-taking due to differences in gender,
education, income, family background, wealth and the broader environment.
The aggregated economic and social effects of this demographic heterogeneity
must therefore be included in any analysis.

In this report, we select ten advanced countries and consider the implications
of their changing demographics for labour markets, consumption patterns,
savings rates and retirement systems. We have long believed that the standard
age-based definitions for assessing an individual’s life cycle are outdated. The
growth of multi-generation families and multi-stage life cycles requires a rethink
of the classic three-stage models of academics P.A. Samuelson, F. Modigliani and
A. Ando2 in the 1950s and their application by actuaries.

The three stages of the conventional model are: 0–14 years (young, non-workers),
15–64 years (working age) and 65+ years (retirees). This was appropriate up
until the 1970s when the number of retirees was still small and the average
retirement period below 15 years. Medical and health advances have since
propelled longevity to historically unprecedented levels. At the other end of the
life cycle, before the early 1970s, most people did not pursue higher education.
In today’s better educated, technologically adept world, people in advanced
countries (and in many developing countries) enter the labour force in their
early twenties, or mid-twenties in the more advanced Nordic countries.

Such demographic and behavioural shifts are triggering the need for new life
cycle models and ways to fund retirement. As a result, policy and investment
research3 has recommended abolishing mandatory retirement ages and
introducing flexible retirement. This is now happening in some advanced
countries and in some progressive emerging countries. In this report, we examine
the underlying demographic data in these countries, how traditional models
should change to reflect modern reality and other ways of ensuring that pension
systems are sustainable.

State Street Global Advisors                                                         2
Demographics & Retirement Research | Global Demographics & Retirement Implications

1. The Evolving Demographic Landscape
One of the major challenges for modern retirement provision is that standard
frameworks have not kept pace with the reality of demographic changes. People
are adapting as their environments become increasingly technical, globalised
and heterogeneous, but neoclassical economic and financial models continue
to be based on old-style assumptions of representative rational consumers
and investors. However, advances in the fields of economics and finance
are leading to the acceptance of not perfectly rational behaviour, as well as
demographic heterogeneity.

In this section, we examine how the demographic landscape has evolved in the
ten countries we have chosen: the US, the UK, Germany, France, Italy, Japan,
Sweden, the Netherlands, Canada and Australia. In later sections, we consider
the impact of this evolution on their retirement systems. The first point to note
is that, although all of these economies are advanced, their core demographic
indicators are not uniform (see Figure 1). This reflects differences in both
individual behaviour and pension, healthcare and welfare systems which
contribute to life expectancy, fertility rates, living standards and population.

Figure 1: Changes in Core Demographic Indicators
Life Expectancy at Birth                                                      Median Age

Australia                                                                     Australia
Canada                                                                        Canada
France                                                                        France
Germany                                                                       Germany
Italy                                                                         Italy
Japan                                                                         Japan
Netherlands                                                                   Netherlands
Sweden                                                                        Sweden
UK                                                                            UK
US                                                                            US
              74        76        78          80            82     84   86                  30             35            40            45     50
                                                Years                                                                   Years
n 1985–1990     n 2015–2020                                                   n 1985         n 2015
Population Growth Rate (annual average)                                       Total Fertility Rate
Australia                                                                     Australia
Canada                                                                        Canada
France                                                                        France
Germany                                                                       Germany
Italy                                                                         Italy
Japan                                                                         Japan
Netherlands                                                                   Netherlands
Sweden                                                                        Sweden
UK                                                                            UK
US                                                                            US
              -0.5            0        0.5            1.0        1.5    2.0                 1.30           1.50           1.70         1.90   2.10
                                             % p.a.                                                               Children per Woman
n 1985–1990    n 2015–2020                                                    n 1985–1990    n 2015–2020

Source: UN, SSGA Demographics.

State Street Global Advisors                                                                                                                         3
Demographics & Retirement Research | Global Demographics & Retirement Implications

Due to the core demographic differences between these countries, it is important
to avoid grouping them into single bloc which could lead to a simplistic analysis.
Instead, we observe the following:

Life Expectancy at Birth: While average life expectancy at birth has
increased by six years since 1985 across these countries, individual nations vary
significantly. Life expectancy in Australia has increased the most from 76.2 years
in 1985 to 83.2 years in 2015. In contrast, the US has the lowest life expectancy
at birth (79.6 years in 2015) and has seen the smallest increase from 74.9 years
in 1985.

Median age: The average Japanese (46.3 years), Italian (45.9 years) and German
(45.9 years) are almost 10 years older than the average US person (37.6 years).

Population growth: Australia and Canada have high annual population
growth with 1.3% and 0.9% for the period 2015–2020, while Japan and Italy
show negative growth with -0.2% and -0.1% respectively.

Fertility rates: Since 1985, total fertility rates have been low across all
countries. Japan, Germany and Italy have the lowest rate at 1.5 children per
woman. However, the changes in fertility rates have not been unidirectional
across all countries, reflecting the impact of national policies and institutions
on individual behaviour.

State Street Global Advisors                                                         4
Demographics & Retirement Research | Global Demographics & Retirement Implications

2. Changing Age Boundaries for Dependency Ratios
Given the changes in life expectancy and median ages presented above, the
standard definitions of labour market entry at 15 and exit at 64 appear to be
out of date. So what would happen if we changed them to reflect the current
demographic reality? Figure 2 demonstrates the impact of different working
age ranges on old-age dependency ratios (i.e., the size of the working age
population versus the size of the elderly population they support).

We compare the old-age dependency ratios using the following age boundaries:

• 65+/15-64: the traditional ratio
• 65+/25-64: captures delayed labour force entry as in the Nordics
• 70+/20-69: reflects the type of ranges seen in Japan, Mexico and Korea
• 70+/25-69: the most realistic ratio over the next ten years
• 75+/25-74: a likely outcome if the quality of health beyond 65 years improves
  everywhere allowing most people to be physically capable of working until
  74; this could be feasible if more people start living into their 90s and beyond

Figure 2: Varying Age Boundaries and Old-Age Dependency Ratios
Australia                                                                      Canada
Number of Dependents per 100 Working Age                                       Number of Dependents per 100 Working Age
30                                28                                           31
                                                                                                                 29
                                                                                                24
25               23                                                            26
                             20                                                                             20
20                                                           18                21                                                           18
                                                16                                                                             16
            15
                                                                                           14
15                                                                             16
                                                        12                                                                             12               11
                                           10                            11                                               10
10                                                                             11
                                                                    6                                                                              7

 5         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74    6         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74

n 1980     n 2015                                                              n 1980     n 2015

France                                                                         Germany
Number of Dependents per 100 Working Age                                       Number of Dependents per 100 Working Age
39                                37                                           44
                                                                                                                 38
34                                                                             39
                 30                                                                             32         32
                            29
                                                                               34
29
                                                                                                                                            26
                                                             23                29                                              24
24          22                                                                             24
                                                21
                                                        18                     24
                                                                                                                                       19
19                                         16                                                                             17                            16
                                                                         15    19
14                                                                  10         14                                                                  10
 9         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74    9         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74
n 1980     n 2015                                                              n 1980     n 2015

Source: UN, SSGA Demographics.

State Street Global Advisors                                                                                                                                  5
Demographics & Retirement Research | Global Demographics & Retirement Implications

Figure 2: Varying Age Boundaries and Old-Age Dependency Ratios (cont’d)
Italy                                                                          Japan
Number of Dependents per 100 Working Age                                       Number of Dependents per 100 Working Age
42                               41                                            53                               50
                 35                                                                             43
36
                                                                               43
31                          27                               27
                                                25                                                                             29           31
26                                                                             33
            21
21                                                                             23                                                                      19
                                                        16               17                                17
                                           14                                              13
16
                                                                                                                           9           10
                                                                               13
11                                                                  8                                                                              5
 6         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74    3         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74
n 1980     n 2015                                                              n 1980     n 2015
Netherlands                                                                    Sweden
Number of Dependents per 100 Working Age                                       Number of Dependents per 100 Working Age
36                                                                             40                               39
                                 34

31               27                                                            35
                                                                                                31         32

26                                                                             30
                            23                                                             25
                                                             20                                                                             24
21                                              18                             25
            17                                                                                                                 21      20
                                                                                                                          18
16                                                      14                     20
                                           12                            12
                                                                                                                                                       14
11                                                                  8          15
                                                                                                                                                   11
 3         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74   10         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74
n 1980     n 2015                                                              n 1980     n 2015
UK                                                                             US
Number of Dependents per 100 Working Age                                       Number of Dependents per 100 Working Age
36                               35                                            29
                            31                                                                                  28
                                                                                                           25
32                                                                             26
                 28
                                                                                                22
28                                                                             23
            23
24                                                           22                20
                                                20      19                                 18                                               17
20                                                                             17
                                           16                                                                                  15      15
                                                                                                                          13
16                                                                       13    14
                                                                    10                                                                                 10
12                                                                             11
                                                                                                                                                   8
 8         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74    8         65+/15-64       65+/25-64       70+/20-69   70+/25-69   75+/25-74
n 1980     n 2015                                                              n 1980     n 2015
Source: UN, SSGA Demographics.

While all of the age boundary definitions illustrate an increase in ageing since
1980, the conventional old-age dependency ratio of 65+/15–64 indicates the
most dramatic rise. Within that bracket, Japan stands out as the country with
the highest dependency ratio. The number of people aged 65+ for every 100
people of working age was 43 in 2015, more than triple the 13 in 1980, reflecting
longevity increases and fertility rate declines. The corresponding numbers for
EU countries have also increased and are now very high. In 2015, Italy, Germany,
Sweden and France respectively had 35, 32, 31 and 30 elderly dependents
supported by every 100 people of working age.

State Street Global Advisors                                                                                                                                  6
Demographics & Retirement Research | Global Demographics & Retirement Implications

However, if the retirement age is increased from 65 years to 70 and 75 years,
the number of elderly dependents per 100 workers falls considerably, even if
the age at which people start working rises from 15 years to 20 and 25 years.
By modifying the age brackets to use more realistic assumptions, the
dependency numbers could be halved (e.g., the dependency ratio in Germany
would fall from 32 to 16 elderly dependents per worker and in the US would fall
from 22 to 10). Even countries with younger populations such as Canada and
Australia could benefit — they have lower dependency ratios but have seen them
rise more quickly in recent decades.

Assuming that everything else remains unchanged, the latter indicates a much
lower burden on governments and younger generations. This would require the
creation of more jobs to accommodate those in the modified working age groups,
a more likely possibility in a period of economic growth.

Labour Force Participation – Male and Female
The previous section showed how dependency ratios adjust with changing
patterns of labour force participation, particularly later entry into the workforce
and later retirement. In this section, we review how participation trends have
changed over time. We consider labour force participation rates in five advanced
countries: Japan, the US, the UK, Germany and France. Figure 3 shows how
overall participation has risen or declined moderately since 1985.

Figure 3: Total Labour Force Participation — 1985 v. 2015
%

66                               65

64                   63                    63   63

62
           60                                           60        60
60                                                                     57

58
                                                                             56
56                                                           55

54              UK                    US        Japan        Germany    France
 1985    2015

Source: UN, SSGA Demographics.

We then break these figures down between male and female. We use male data
to show trends within age groups and across different countries only because
historically there have been more men than women in the workforce.

State Street Global Advisors                                                          7
Demographics & Retirement Research | Global Demographics & Retirement Implications

     Male
     Over the last 30 years, there has been an almost universal decline in the number
     of men participating in the labour force. In all countries (except Japan), the
     sharpest drop in participation has been in the youngest age group (15–24 years).
     This was mainly driven by more years in education, an economic shift from
     agriculture to manufacturing and services and higher agricultural productivity.

     Figure 4 shows that the US has experienced the biggest decline in labour force
     participation rates for this age group from 73% (1985) to 56% (2015), followed
     by France from 55% (1985) to 40% (2015) and the UK from 74% (1985) to 60%
     (2015). Labour market shifts have led employers to seek out first-time employees
     with higher levels of education and skills than in previous generations.

     Figure 4: Male Labour Force Participation Rates (across different age groups)
     Men aged 15–24 years                                                  Men aged 25–34 years
     %                                                                     %
     75                                                                    100
               74             73                                                                                              97
                                                                                                          97   96
     65                                           61                                 95         95
                     60                                                                    93
                                   56                                       95
                                                            55
                                                                                                                    92             92
     55                                                51                                                                90
                                                                                                     89
17                                      44                            17    90
     45                                      43
                                                                 40
14                                                                    14

11   35             UK         US       Japan     Germany   France    11    85            UK     US       Japan     Germany   France
     n 1980   n 2015                                                       n 1980   n 2015
     Men aged 35–44 years                                                  Men aged 45–54 years
     %                                                                     %
     100                                                                   100
                                        98                  98
                                                                                                          97 96
                                                  97
               96                            96                                                                     95
                              95                       95
      96                                                                    95
                                                                 94                  93                                  93   93
                     93                                                                         91                                 91
                                                                                           90
17    92                                                              17    90
                                   90
14                                                                    14
                                                                                                     86
11    88            UK         US       Japan     Germany   France    11    85            UK     US       Japan     Germany   France
     n 1980   n 2015                                                       n 1980   n 2015
     Men aged 55–64 years                                                  Men aged 65 years and older
     %                                                                     %
     90                                                                    40                             39
                                                                                                               31
     80                                 85 86          75
                     71                                                    30
               69                  70
                              68                                                                     23
     70
                                                                           20                   16
                                                  59                                       14
     60                                                          55
                                                                                     8                                   9
                                                                           10                                        6
     50                                                     44                                                                4    4

11   40             UK         US       Japan     Germany   France    11    0             UK     US       Japan     Germany   France
     n 1985   n 2015                                                       n 1980   n 2015

     Source: ILO, SSGA Demographics.

     State Street Global Advisors                                                                                                       8
Demographics & Retirement Research | Global Demographics & Retirement Implications

     The labour force participation rates for other working age groups including 25–34,
     35–45 and 45–54 years have also decreased across all these countries since 1985,
     although the declines have been less steep than that of age group 15–24. Again, the
     most significant change occurred in the US across all groups. In particular, the US
     labour force participation rate for 25–34 year olds has dropped by 6% from 95% in
     1985 to 89% in 2015 while in Germany and Japan the decline was only 2%.

     In contrast, the participation rates of those in the oldest working age group
     (55–64) have increased in all countries, most dramatically in EU countries such
     as Germany (from 59% in 1985 to 75% in 2015) and France (from 44% in 1985
     to 55% in 2015).

     As advocated here and elsewhere,4 there are also more people working beyond
     the traditional retirement age. For example, the labour force participation rate
     for men aged 65+ in the US has risen by 7% from 16% (1985) to 23% (2015). There
     have been similar increases in the UK (6%) and Germany (3%) over the last three
     decades. Given improvements in healthcare and working conditions, and the shift
     from manufacturing to services, this trend is likely to continue across developed
     and some developing countries.

     Female
     While male labour participation rates have declined sharply over the last 30
     years, female participation rates have increased (Figure 5). In Germany, for
     example, the 14% increase in female participation has offset the reduction in male
     labour force participation and helped increase the total rate from 55% in 1985 to
     60% in 2015. Moreover, across the G7 countries, the number of female graduates
     is now higher than that of males, and in most countries, women now outlive men
     by three to five years. Because of these factors, we expect this trend to continue,
     especially as the gender bias in favour of men diminishes.

     Figure 5: Male v. Female Labour Force Participation Rates
     Total Male Labour Force Participation                                 Total Female Labour Force Participation
     %                                                                     %
     80                                 79                                 60
                              76                                                           57        57
                                                                                                55                        55
     75        73                                                          55
                                                  71                                                                                52
                                   69        70                                                                50
                     69                                     69
     70                                                                    50        48                   48
                                                                                                                               47
                                                       66
     65                                                                    45
                                                                 61                                                  41
11   60             UK         US       Japan     Germany   France    11   40             UK     US       Japan      Germany   France
     n 1985   n 2015                                                       n 1980   n 2015

     Source: ILO, SSGA Demographics.

     State Street Global Advisors                                                                                                        9
Demographics & Retirement Research | Global Demographics & Retirement Implications

3. Macro Implications of Demographic Changes
This section examines the impact of demographic shifts on economies and
pensions. It looks at changes in the way people consume and save and how
they are adapting to increasing longevity.

Consumers and Savers5
An individual’s consumption and savings patterns are affected by a number
of characteristics: age, gender, country, education, health, family background
and environment. In addition, globalisation and economic progress have altered
these patterns across the world.

Past research has tended to document the changing nature of consumers and
savers6 as it affects aggregate GDP and capital flows.7 These days, more micro
economic trends are observable. Not only are people living longer and having
fewer children but their behaviour is changing too.8 While total consumption
expenditure in advanced countries accounts for a relatively similar proportion
of GDP (60+%) compared to 20 years ago, underlying spending patterns exhibit
much greater divergence, as demonstrated by changes in the four elements of
consumption expenditure: non-durables, semi-durables, durables and services.
These components exhibit different elasticities with respect to incomes and vary
over the business cycle (e.g., spending on durables tends to fall during recessions,
see Figure 11 in Appendix).

Figure 6 shows how these components have changed over time (for illustrative
purposes, we show only US and Canada data). Services expenditure accounted for
the largest share of household consumption in both countries in 2016, followed
by that of non-durables, durables and semi-durables. However, compared to 1985,
the shares of services and non-durables have reduced significantly. Spending by
different age groups also varies in terms of relative consumer expenditures on
clothing, food, healthcare, transport etc. and across countries (see Figure 12 in
the Appendix).

State Street Global Advisors                                                           10
Demographics & Retirement Research | Global Demographics & Retirement Implications

Figure 6: Household Consumption Expenditure by Component (2016 versus 1985)
US                                                                                       Canada
Share of Total Consumption Expenditure (%)                                               Share of Total Consumption Expenditure (%)
100                                                                                      100

80                                                                                       80                         48.4                               53.6
                           67.0                                  63.6
60                                                                                       60

40                                                                                       40                                                            23.0
                                                                     17.6                                           33.4
                           23.5
20                                                                   7.9                 20                                                            8.5
                                                                                                                     7.6
                            4.7                                  11.0                                               10.6                               14.9
                            4.8
  0                        1985                                  2016                      0                        1985                               2016
n Durable goods       n Semi-durable goods      n Non-durable goods         n Services   n Durable goods       n Semi-durable goods   n Non-durable goods     n Services

Source: OECD, SSGA Demographics.

Trends in household savings rates too vary over time and by country, but are
typically trending lower (Figure 7). Some countries such as the US and the
UK have experienced greater volatility than others such as Germany and
France where savings rates tend to be higher. We believe that all countries
need to increase household savings to help defray longevity-related health
and pension expenditures. However, there is a trade-off between savings and
consumption. The more people save, the less they tend to consume, leading
to lower GDP. Changes in consumption and savings patterns also affect capital
flows, inflation and the current accounts and fiscal positions of countries with
open economies.

Figure 7: Household Savings Trends as % of GDP
% of Household Disposable Income
20

15

10

 5

 0

-5       1985                1990             1995            2000               2005    2010                2015

— Canada        — France     — Germany       — Japan   — UK    — US

Source: OECD, SSGA Demographics.

State Street Global Advisors                                                                                                                                               11
Demographics & Retirement Research | Global Demographics & Retirement Implications

Employment and Labour Force
As longevity has increased and life cycles have changed, individuals and families
have changed their work patterns, while still responding to changing incentives
over the business cycle.

Lifelong learning, changing skill requirements and the absence of guaranteed
lifetime employment are resulting in individuals working for multiple employers
in a variety of roles. In addition, more people are able to work flexibly, remain
employed past official retirement ages, embrace self-employment and combine
different jobs at the same stage of their life cycle to generate the income they
require. At the same time, globalisation, immigration and technological
advances (especially automation) are changing the workplace landscape,
compelling individuals and governments to rethink traditional models of
employment. See Figure 13 in Appendix showing changes in self-employment
and part-time working.

The effective retirement age is when people actually retire on average.
Figure 8 shows that people retire earlier than their pensionable age in countries
such as France, Germany, Netherlands and Italy. The reason for this is explained
in Section 4. In the UK and Australia, labour market exit occurs close to official
retirement age, while late retirement is the norm in countries such as Japan,
Canada, Sweden and the US. For women, late retirement is most common in
Japan where the effective retirement age is significantly higher than the official
retirement age (by 3.8 years).

Figure 8: Official versus Effective Retirement Age, 2016
Age (years)
71

69

65

63

61

59      Japan    Canada    Sweden      US      Australia    UK      France   Germany Netherlands   Italy

n Official Retirement Age, Men     n Official Retirement Age, Women
n Effective Retirement Age, Men   n Effective Retirement Age, Women

Source: OECD, SSGA Demographics.

State Street Global Advisors                                                                               12
Demographics & Retirement Research | Global Demographics & Retirement Implications

Pensions at the Aggregate Level
Pensions and retirement savings are affected by what individuals as consumers,
savers and workers do in the pre-retirement stages of life. When taken in
aggregate, these can have a big impact on government budgets. Across some
OECD countries, pensions and health expenditures now account for at least
60% of all public social spending.9 Countries with older populations like
Germany, France, Italy and Japan have much higher spending on pensions and
health compared to those such as Australia and Canada. Fiscal, labour and
monetary policy all need to respond to these changes.

Figure 9: Public Social Expenditures as % of GDP

35

30

25

20

15

10

 5

 0    Australia   Canada      France    Germany      Italy      Japan   Netherlands Sweden   UK   US

 Old Age     Health       Others

Others include Family, Incapacity, Labour Market, Housing and Other Social.
Source: OECD, SSGA Demographics.
The data relates to 2013, except for Australia and Canada which relates to 2014.

State Street Global Advisors                                                                           13
Demographics & Retirement Research | Global Demographics & Retirement Implications

4. How Retirement Systems Can Adapt
to Demographic Changes
The earliest national and corporate pension plans in developed countries were
DB in nature, with risks borne by the plan sponsors who were companies or
governments. By the late 1970s and early 1980s, issues with DB plans led to the
emergence of DC schemes, where risks were passed to plan participants. The
move towards DC and away from DB is an acknowledgement that past promises
made by DB pension systems are not sustainable given changing demographic
and behavioural trends.10

Another factor behind this pension problem11 has been the steady decline in
retirement ages from 1970 until the late 1990s. This coincided with a period
of high economic growth combined with higher asset returns (in the 1980s
and 1990s) that allowed many workers to retire early or work less. As a result,
early retirement became the norm in European countries such as Italy, France,
Germany and the Netherlands (see Section 3), a trend that only started to
reverse in 2002 (Figure 10).

Figure 10: Changes in the Effective Retirement Age in OECD Countries
Effective Retirement Age
 70

 68

 66

 64

 62

 60      1971              1980           1989   1998       2007       2016
— Men      — Women

Source: OECD Pensions at a Glance 2017.

Many countries have since introduced reforms12 to lift the minimum retirement
age. However, these increases are too modest to have an immediate impact on the
sustainability of most retirement systems. On average, the normal retirement age
in OECD countries will increase by 1.5 years for men and 2.1 years for women,
reaching just under 66 years in 2060.13 During this same period, the UN projects
that life expectancy for a 65 year old in the US will increase by 3.2 years.14
Planned changes alone therefore will not ensure retirement systems are

State Street Global Advisors                                                         14
Demographics & Retirement Research | Global Demographics & Retirement Implications

sustainable and working lives will have to extend further. Holistic reforms
in labour markets, tax, education and health are essential to mitigate the strains
of fiscal sustainability and will need to vary across countries.15

System design, therefore, is crucial. We have seen how using modified definitions
of working ages leads to far more reasonable old-age dependency ratios. If most
countries raised their effective retirement ages by five years, they could reduce
pension liabilities by about 100% of GDP.16 Retiring later also has a positive
impact on the expected outcomes for individuals: it reduces the expected
retirement period and allows participants to save more and prolong their
investment horizons.

The challenge for policymakers is to find ways to support the transition towards
a more sustainable retirement.17 A recent survey found that almost two-thirds of
EU citizens would prefer to combine a part-time job with a partial pension than
to fully retire; yet in Europe only about 10% of individuals aged 60–69 currently
combine work and a pension.18,19 Many retirement systems discourage working
while drawing a pension, e.g., by reducing the pension payout after a certain
income level. Extending working lives will therefore require better incentives
for working full or part-time beyond current retirement ages, as well as
encouragement to continue saving or to preserve accumulated assets for
later drawdown.20

We discuss below some policy changes that we believe would make retirement
systems more sustainable in an aggregate sense.

A. Abolish Mandatory Retirement Ages
One way to encourage people to extend their working lives is to abolish mandatory
retirement ages.21 This has already been done in countries such as the UK, with
subsequent increases in age 65+ workforce participation. As well as allowing
people to work longer, this change helps reduce the anchoring effect of people
thinking that passing a particular age means that they should give up work.

Another important question is how employers will facilitate and value
these older workers. While some employers already see them as a source of
competitive advantage, many are unprepared for the rise in an older workforce.22
Policymakers may need to encourage the hiring of older workers, e.g., by
subsidising healthcare costs, reducing employment protections and offering these
workers access to lifelong learning and re-training.

State Street Global Advisors                                                         15
Demographics & Retirement Research | Global Demographics & Retirement Implications

B. Increase State Pension Flexibility
The state benefit is an important part of the retirement savings system in most
countries. Although the precise format varies, most “pillar one” (basic state
pension) systems incentivise workers to defer retirement by providing higher
guaranteed benefits if they delay making a claim. Many countries utilise an age
band for claiming benefits, with a minimum eligibility age and a maximum age
for when benefits have to start. While it is possible to combine work and pensions
after the normal retirement age in most OECD countries, earning wage income
can often reduce the pension entitlement, providing a disincentive to prolonging
employment — though some countries are now taking steps to tackle this.
Moreover, many countries operate a cliff edge system whereby the benefit must
be taken in full by a certain age. For example, US Social Security payments have
to start at age 70 and there is no option for claiming a partial benefit to facilitate
‘partial retirement’.

We believe a more flexible pillar one system could encourage people to work for
longer. In Sweden, people who have reached the minimum age of eligibility (62)
can take 25%, 50%, 75% or 100% of their state benefit and modify the percentage
when desired at an actuarially fair rate. There is also no maximum age by which
full payments have to start.23 Such flexibility may be particularly useful for
people who have started saving for retirement later in life or are physically unable
to work full time but for whom part-time work is feasible. These systems may
be more complex to administer and will need to reflect different types of taxable
income and incentives in order to avoid unintended consequences.

We would also recommend removing upper age limits for claiming pillar one
entitlements, so that people can make best use of the cost-efficient longevity
insurance these provide. “Pillar two” occupational pension provision (often but
not always private pension schemes) in most countries is moving away from DB
towards DC. A major drawback of DC relative to DB is the lack of guaranteed
lifetime income. If individuals could choose when to claim the state benefit, they
could use the pillar one benefits as a longevity backstop when their pillar two
funding ran out, providing a main source of income late in life, rather than a
steady source of income throughout retirement.

In Australia, for example, eligibility for the Age Pension is based on an asset
test (reassessed annually) rather than age. Retirees are not eligible for the Age
Pension until they have drawn their assets down to a minimum level, after which
they receive the flat rate Age Pension for the rest of their lives. Economist Michael
Johnson suggests a similar modification to the UK State Pension, postponing the
start of payments to age 80 but doubling the payment.24

State Street Global Advisors                                                             16
Demographics & Retirement Research | Global Demographics & Retirement Implications

To return to our example of US Social Security, we calculate that if it were
possible to defer Social Security payments until age 75 at an actuarially fair
rate, this would increase replacement rates (the proportion of pre-retirement
income replaced by the state benefit) by as much as 50%.25 The ability to postpone
payments until a later date and receive higher benefits for a shorter period could
be particularly valuable for higher income earners, who would then receive
meaningful income and valuable longevity insurance from the state pension.

C. Incorporate Income Differences into Retirement Policy
While most DB plans are proving unsustainable amid changing demographics,
a move to a pure DC world could exacerbate income inequality. Those on lower
incomes might have to work much longer than those on higher ones before they
felt they had saved enough to retire on. Most pension reforms currently under
way (such as in the Netherlands or Germany) aim to introduce more elements
of DC to replace legacy DB promises. However, a shift to DC does not reduce
dependency on the state pension; quite the contrary, the state pension plays
an even more important role in a DC than a DB world as it is the only source
of guaranteed lifetime income.

State pension entitlements are typically flat rate (e.g., in the Netherlands,
the UK or Australia) or provide low income earners with a relatively high
replacement rate (e.g., the US). In both cases, they represent an important
element of income redistribution. Reaching a 45% replacement rate based
only on DC savings would require a consistent savings rate of roughly 15%,
which may be a struggle for very low earners.

Unfortunately, most efforts to improve the sustainability of pillar one provision
focus on increasing the age of eligibility. But this can be disproportionately unfair
to low earners who rely most on receiving the state pension, tend to have lower
than average life expectancy and are more likely to be employed in professions
in which working lives are harder to extend.26 Many low income employees also
start working earlier than those with higher levels of education. Requiring all
employees to extend their working lives to the same minimum retirement age
would therefore reduce the ratio of time spent in retirement relative to work
specifically for lower income employees and would not be equitable.27

To be fair to all income groups, changes to pension systems will have to take
into account the increasing heterogeneity in terms of life expectancy and savings
capacity between income cohorts. One solution could be to link the minimum
eligibility for full retirement benefits to years of work rather than age; low income
individuals who started work before university graduates would then qualify for

State Street Global Advisors                                                            17
Demographics & Retirement Research | Global Demographics & Retirement Implications

full retirement benefits at an earlier age, compensating for lower life expectancy.
Such an approach was considered by the UK review of state pension age led by
John Cridland, former Director-General of the Confederation of British Industry,
but no recommendation to that effect was made.28

D. Incentivise Higher Savings — Behavioural Nudges & Auto-Enrolment
The measures outlined above go some way towards improving the sustainability
of retirement systems. However, a healthy system still relies on adequate
savings from the working population. Recent evidence shows that if mandating
retirement savings is politically impossible, a variety of behavioural interventions
(‘nudges’) can be used to help people save more for their retirement.

One such nudge is automatic enrolment. Since it was introduced in the UK,
opt outs have been only c. 5%–15%.29 Auto-enrolment is also used voluntarily by
many large US DC plans and is under consideration for improving participation
in Ireland. Australia goes further with compulsory participation for all employees
in a retirement plan. ‘Save more tomorrow’ approaches — involving automatically
escalating contribution rates — can also boost contributions over time.30 These
allow participants to be introduced to the plan at a low rate, avoiding any adverse
reaction to reductions in take-home pay, before being gradually raised to the
required long-term levels. A growing number of large US plans, the UK and the
Australian systems have adopted this or a similar approach.31 However, the UK
and Australian schemes are not structured to apply the save more tomorrow
approach to individual participants beyond the launch cohort.

Governments considering introducing auto enrolment regimes should think
carefully about how high to set the default enrolment rate, as setting too high
a rate could cause low income earners to opt out. On the other hand, a rate that
is appropriate for low income cohorts will be too low for those in higher wage
groups. One possibility could be to have different auto-enrolment rates for
different income cohorts — again, policy must take into account the increasing
heterogeneity of the different population groups.32 Matching can also be used
to encourage voluntary contributions. Employers can offer to match employee
contributions or tax relief on employee contributions can be presented as a form
of matching (relief from 20% tax can be recast as 1:4 matching, which may be
simpler to understand).

State Street Global Advisors                                                           18
Demographics & Retirement Research | Global Demographics & Retirement Implications

Many countries do not allow workers any access to their savings prior to
retirement. In countries such as the US where employees are able to access
the money, reducing pre-retirement ‘leakage’ of pension assets should improve
effective savings rates. This can include discouraging early withdrawals and
ensuring accumulated assets are rolled over into a retirement plan rather than
cashed out when people switch jobs. An example where the reverse has happened
is the UK. From 2015, individuals were no longer required to buy an annuity
on retirement and had full access to retirement assets from the age of 55. Early
evidence shows a significant number of participants taking withdrawals in their
50s for non-retirement reasons (e.g., leisure, home improvement) with relatively
little consideration for the impact on longer term retirement income.33

E. Introduce Default Solutions for the Retirement Phase
When saving rates are low relative to required outcomes, it is obviously
important to maximise the efficiency of those savings. Unfortunately this does
not always happen. When DC was initially introduced in the US, employees had
the freedom to make their own investment choices. This often led to sub-optimal
outcomes, with employees investing in vehicles with inappropriate risk profiles
for their ages. There is little evidence that encouraging participant investment
choice leads to better outcomes and significant evidence to the contrary.34 Most
DC plans now allow employee contributions to be automatically directed to well-
governed default funds, such as target date funds, which provide participants
with age-appropriate levels of risk at relatively low ‘institutional’ fee rates.

While the accumulation phase of DC savings is well understood, most DC savings
vehicles do not smoothly transition into providing employees with an income
stream in retirement. Instead, in countries such as Australia, the US and the UK,
DC savers are typically presented with a large lump sum that has to last for the
rest of their lives. This introduces the potential for behavioural biases such as loss
aversion and a disinclination to annuitise a least a portion of the sum to protect
against longevity.

Indeed, experience from across the world shows that people are often reluctant
to buy an annuity unless they are legally required or strongly incentivised
to do so. Some use investment-led ‘drawdown’ approaches with no longevity
insurance, but then need to self-insure against living to older ages. Effectively,
this implies that they have to underspend during retirement in order to protect
themselves against the risk of living to a very old age and, as a consequence, leave
unintended bequests.

State Street Global Advisors                                                             19
Demographics & Retirement Research | Global Demographics & Retirement Implications

Moreover, because the institutional savings vehicles used during the
accumulation phase do not typically provide employees with an income when
they retire, most employees move their accumulated savings into a retail vehicle.
This means that they usually pay higher fees and may receive little or no advice
as to how to draw down their assets, which can have a significant impact on
their income.

Fortunately, new solutions are emerging that allow employees to opt for a default
vehicle on retirement, flexible access to savings (possibly in concert with part-
time work) and a deferred annuity that provides secure income from, say, age 80,
as well as mortality pooling which removes the need for self-insurance. Greater
awareness of these types of solutions should help retirees better manage their
longevity risk.

State Street Global Advisors                                                         20
Demographics & Retirement Research | Global Demographics & Retirement Implications

Conclusion
People are living longer and dependency ratios based on the traditional three-
stage life cycle appear out of date and unsustainable. However, if we modify the age
range of the working population to better reflect the modern demographic reality,
defined as people aged 20–69 rather than 15–64, dependency ratios begin to look
more manageable and pension systems more sustainable.

Nonetheless, further policy changes are required to ensure projected increases
in the retirement age keep pace with projected increases in life expectancy.
The transition towards a more DC-based retirement saving system should help
accelerate the adjustment to longer life expectancy. But it also risks exacerbating
income inequality as lower income cohorts typically have lower life expectancy
and are less able to extend their working lives.

In a nutshell, a sustainable retirement system with feasible dependency ratios
requires that workers save more and retire later. There are a number of policy
actions for individuals, employers and national retirement systems that
we recommend below.

1. Abolish mandatory retirement ages and promote flexible retirement patterns
2. Incentivise women to enter the workforce (via education, pension and childcare
   reforms and technology)
3. Promote lifelong learning by increasing investment in re-training programs for
   older workers
4. Create flexible pillar one systems that encourage later life working patterns such
   as combining part-time work with partial retirement
5. Abolish the upper limit on claiming pillar one entitlements to encourage delay
   in take-up
6. Link pillar one eligibility to the number of years worked to ensure fair treatment
   of lower income cohorts
7. Ensure consistent working life contributions to an appropriate retirement
   savings vehicle by mandate where possible or by nudges (e.g., auto-enrolment)
8. Integrate default solutions for the retirement phase into the default savings
   vehicle, providing a smooth transition from accumulation phase into retirement
9. Maximise efficiency of retirement savings by incorporating longevity protection
   into default solutions for the retirement phase

For the success of the recommendations above, it is important for asset owners
and managers, alongside policy makers, to enhance their understanding of the
impact of demographic changes on economies and households in order to provide
new savings and investment solutions for workers and retirees at different stages
of the lifecycle.

State Street Global Advisors                                                            21
Demographics & Retirement Research | Global Demographics & Retirement Implications

Appendix
Figure 11: Household Consumption Expenditure Annual Growth by Component (1985–2015)
US                                                                                                             Canada
YOY % (constant USD, base year = 2009)                                                                         YOY % (constant CAD, base year = 2007)
15                                                                                                             15

10                                                                                                             10

 5                                                                                                              5

 0                                                                                                              0

 5                                                                                                              5

-10      1985           1990           1995           2000            2005         2010        2015            -10       1985           1990          1995      2000      2005          2010        2015

— Durable Goods         — Non-durable Goods            — Semi-durable Goods            — Services              — Durable Goods          — Non-durable Goods      — Semi-durable Goods      — Services

Source: OECD, SSGA Demographics.

Figure 12: Household Expenditure by Age Group (2016)
US                                                                                                             Canada
% of Average Annual Expenditure                                                                                % of Average Annual Expenditure
100                                                                                                            100

 80                                                                                                             80

 60                                                                                                             60

 40                                                                                                             40

 20                                                                                                             20

  0
Demographics & Retirement Research | Global Demographics & Retirement Implications

1
   (a) Peter Drucker (1976), The Unseen Revolution,(b) European Commission                28
                                                                                             Cridland, J (2016) ‘State Pension age independent review: interim report’ HMSO
   Sustainability Report (c) OECD Pensions at a Glance and (d) Credit Suisse (2015),         London 2016.
   EU’s evolving Demographics & Pensions Needs attention.                                 29
                                                                                             DWP (2014) Automatic enrolment opt out rates: findings from qualitative research
2
   P.A. Samuelson (1958), F. Modigliani & Brumberg (1954) A. Ando made pioneering            with employers staging in 2014, Department for Work and Pensions, HMSO London.
   contributions to understanding consumption and saving behavior of individuals.         30
                                                                                             Benartzi, S and Richard Thaler. 2007. ‘Heuristics and Biases in Retirement Savings
3
   “The Demographic Manifesto: New Jobs, New People”, G. Keating, R. Hokenson &              Behavior.’ Journal of Economic Perspectives, 21(3): 81-104.
   A. Roy (Credit Suisse Demographics Research, 2000), also a series of OECD Country      31
                                                                                             Vanguard, “How America Saves 2016”.
   reports titled “Live Longer, Work Longer” in 2006 advocating the need to work          32
                                                                                             Reilly & Byrne (2017) “Investing for Retirement in a Low Returns Environment –
   longer as we live longer.                                                                 Making the Right Decisions to Make the Money Last” PRC WP2017-7.
4
   By one of the authors of this report in the Demographic Manifesto (2000) and           33
                                                                                             State Street Global Advisors (2016) ‘New Choices: Big Decisions’ White paper.
   presented to governments across the globe.                                             34
                                                                                             Cronqvist, H., & Thaler, R. (2004). Design Choices in Privatised Social-Security
5
   “Why Demographics Matters? And How?”, A. Roy (Credit Suisse Demographics                  Systems: Learning from the Swedish Experience. The American Economic Review,
   Research, 2006).                                                                          94(2), 424-428.
6
   Credit Suisse Demographics Research (2006) “Changing Consumers and Workers”
   and also Credit Suisse Demographics Research (2015), “Demographic Focus-               References
   Changing Global Consumers”.
                                                                                          Belbase, A., Sanzenbacher G.T., Gillis C.M., (2016) ‘How do Job Skills that Decline
7
   Credit Suisse Demographics Research (2007), “Demographics, Savings and Capital         with Age Affect White-Collar Workers?’ Center for Retirement Research at Boston
   Flows” and Credit Suisse Demographics Research (2009), “Demographics, the              College Working Paper April 2016, Number 16-6.
   Japanese Current Account & A Disappearing Savings Rate”.
                                                                                          Benartzi, S and Richard Thaler. 2007. ‘Heuristics and Biases in Retirement Savings
8
   “Microtrends (2009)” by Mark Penn & Kinney Zalesne and “Who are We? (2004)”
                                                                                          Behavior.’ Journal of Economic Perspectives, 21(3): 81-104.
   by Sam Huntington provide micro and macro-sociological insights into the changing
   nature of consuming individuals.                                                       Credit Suisse Demographics Research (2000), “The Demographic Manifesto
                                                                                          – New Jobs, New People”.
9
   OECD (2016), “Social spending stays at historically high levels in many OECD
   countries”, Social Expenditure Update 2016.                                            Credit Suisse Demographics Research (2006), “Why Demographics Matters?
10
   Peter Drucker (1976), The Unseen Revolution; F Modigliani & A Muralidhar (2005)        And How?”.
   “Rethinking Pension Reform”.                                                           Credit Suisse Demographics Research (2007), “Demographics, Savings and
11
   Author’s calculations, based on OECD data on effective retirement age and UN           Capital Flows”.
   population projections.                                                                Credit Suisse Demographics Research (2009), “Demographics, the Japanese Current
12
   Credit Suisse Demographics Research (2011), “How Increasing Longevity Affects us       Account & A Disappearing Savings Rate”.
   All?” argued for the urgent need for holistic policy reform across pensions, labour,   Credit Suisse Demographics Research (2010), Macro “Fiscal Sustainability” to Micro
   health, insurance and social benefits.                                                 “Economic Conditions of the Old “in the “Oldest Five “Countries.
13
   OECD Pensions at a Glance 2017.                                                        Credit Suisse Demographics Research (2014), “A Demographic View: Do not write
14
   United Nations World Population Prospects: the 2015 Revision.                          off US GDP growth”.
15
   Credit Suisse Demographics Research (2011), Macro “Fiscal Sustainability” to Micro     Credit Suisse Demographics Research (2015), “Demographic Focus-Changing
   “Economic Conditions of the Old” in the “Oldest Five” Countries.                       Global Consumers”.
16
   Healey & Reilly 2015’ “Global Pensions Underfunding” M-RCBG Associate Working          Credit Suisse Demographics Research (2015), “EU’s evolving demographics
   Paper No. 39.                                                                          & pensions need attention”.
17
   ADB Institute conference on Longevity and Pensions speeches converted to ADBI          Cridland, J (2016) ‘State Pension age independent review: interim report’ HMSO
   Working Paper (2012), “Innovative Approaches to Managing Longevity Risk in Asia:       London 2016.
   Lessons from the West”, A. Roy.
                                                                                          Cronqvist, H., & Thaler, R. (2004). Design Choices in Privatised Social-Security
18
   Eurofound, 2016.
                                                                                          Systems: Learning from the Swedish Experience. The American Economic Review,
19
   OECD Pensions at a Glance 2017.                                                        94(2), 424-428.
20
   Gratton L. and Scott A. (2016) The 100 Year Life: Living and Working in an Age of      DWP (2014) Automatic enrolment opt-out rates: findings from qualitative research
   Longevity London: Bloomsbury.                                                          with employers staging in 2014, Department for Work and Pensions, HMSO London.
21
   Credit Suisse (2000), The Demographic Manifesto advocates abolishing fixed
                                                                                          Eurofound (2012), Income from work after retirement in the EU, Publications Office
   retirement ages and move towards flexible retirement in a digital global workplace.
                                                                                          of the European Union, Luxembourg.
22
   Sonsino (2017) “Helping Employers Become Age-Ready” PRC WP2017-11.
                                                                                          European Commission Sustainability Report (2015).
23
   See Pensions Myndigheten – Swedish Pensions Agency for a description of how
   the Swedish system works.                                                              Franco Modigliani & Arun Muralidhar, (2005), Rethinking Pension Reform, Cambridge.
24
   Johnson, M. (2016) ‘The State Pension: No Longer Fit for the Purpose’, Pointmaker,     Gratton L. and Scott A. (2016) The 100 Year Life: Living and Working in an Age of
   Centre for Policy Studies Michael Johnson, Centre for Policy Studies, 2016.            Longevity London: Bloomsbury.
25
   Reilly & Byrne (2017) “Investing for Retirement in a Low Returns Environment –         Healey & Reilly 2015’ “Global Pensions Underfunding” M-RCBG Associate Working
   Making the Right Decisions to Make the Money Last” PRC WP2017-7.                       Paper No. 39.
26
   Belbase, A., Sanzenbacher G.T., Gillis C.M., (2016) ‘How do Job Skills that Decline    Johnson, M. (2016) ‘The State Pension: No Longer Fit for the Purpose’, Pointmaker,
   with Age Affect White-Collar Workers?’ Center for Retirement Research at Boston        Centre for Policy Studies.
   College Working Paper April 2016, Number 16-6.                                         Mark Penn & Kinney Zalesne (2009), “Microtrends”.
27
   Sanzenbacher et al., 2015.                                                             OECD (2006), “Live Longer, Work Longer”.

State Street Global Advisors                                                                                                                                                    23
Demographics & Retirement Research | Global Demographics & Retirement Implications

OECD (2016), “Social spending stays at historically high levels in many OECD             Reilly & Byrne (2017) “Investing for Retirement in a Low Returns Environment
countries”, Social Expenditure Update 2016.                                              – Making the Right Decisions to Make the Money Last” PRC WP2017-7.
OECD (2017), Pensions at a glance OECD and G20 indicators.                               Sam Huntington (2004), “Who are We?”.
P.A. Samuelson (1958), F. Modigliani & Brumberg (1954) A. Ando made pioneering           Sonsino (2017) “Helping Employers Become Age-Ready” PRC WP2017-11.
contributions to understanding consumption and saving behavior of individuals.           State Street Global Advisors (2016) ‘New Choices: Big Decisions’ White paper
Pensions Myndigheten ‘Du bestammer sjalv nar du vill ta ut pension’, Publication of      published by SSGA available on https://www.ssga.com/definedcontribution/uk/.
the Swedish Pensions Authority available on www.pensionsmyndigheten.se.                  United Population Division (2015), World Population Prospects, New York.
Peter Drucker (1976), The Unseen Revolution: How Pension Fund Socialism came to          United Population Division (2017), World Population Prospects, New York.
America, Rheinemann, London.
                                                                                         Vanguard, (2016) ‘How America Saves 2016’ Annual report on trends
Redington Report, “The Age of Responsibility, Economics and Demographics”, (2015).       in retirement saving issued by Vanguard.

ssga.com

For investment professional use only. Not for public use.                                Building, 7th floor Herikerbergweg 29 1101 CN Amsterdam, Netherlands.
                                                                                         T: 31 20 7181701. SSGA Netherlands is a branch office of State Street Global
State Street Global Advisors Worldwide Entities                                          Advisors Limited. State Street Global Advisors Limited is authorised and regulated
Australia: State Street Global Advisors, Australia, Limited (ABN 42 003 914 225)         by the Financial Conduct Authority in the United Kingdom. Singapore: State Street
is the holder of an Australian Financial Services Licence (AFSL Number 238276).          Global Advisors Singapore Limited, 168, Robinson Road, #33-01 Capital Tower,
Registered office: Level 17, 420 George Street, Sydney, NSW 2000, Australia. T:          Singapore 068912 (Company Reg. No: 200002719D, regulated by the Monetary
+612 9240 7600. F: +612 9240 7611. Belgium: State Street Global Advisors Belgium,        Authority of Singapore). T: +65 6826 7555. F: +65 6826 7501. Switzerland: State
Chaussée de La Hulpe 120, 1000 Brussels, Belgium. T: 32 2 663 2036. F: 32 2 672          Street Global Advisors AG, Beethovenstr. 19, CH-8027 Zurich. Authorised and
2077. SSGA Belgium is a branch office of State Street Global Advisors Limited. State     regulated by the Eidgenössische Finanzmarktaufsicht (“FINMA”). Registered with the
Street Global Advisors Limited is authorised and regulated by the Financial Conduct      Register of Commerce Zurich CHE-105.078.458. T: +41 (0)44 245 70 00. F: +41 (0)44
Authority in the United Kingdom. Canada: State Street Global Advisors, Ltd., 770         245 70 16. United Kingdom: State Street Global Advisors Limited. Authorised and
Sherbrooke Street West, Suite 1200 Montreal, Quebec, H3A 1G1, T: +514 282 2400           regulated by the Financial Conduct Authority. Registered in England. Registered No.
and 30 Adelaide Street East Suite 500, Toronto, Ontario M5C 3G6. T: +647 775 5900.       2509928. VAT No. 5776591 81. Registered office: 20 Churchill Place, Canary Wharf,
Dubai: State Street Bank and Trust Company (Representative Office), Boulevard            London, E14 5HJ. T: 020 3395 6000. F: 020 3395 6350. United States: State Street
Plaza 1, 17th Floor, Office 1703 Near Dubai Mall & Burj Khalifa, P.O Box 26838, Dubai,   Global Advisors, One Lincoln Street, Boston, MA 02111-2900. T: +1 617 786 3000.
United Arab Emirates. T: +971 (0)4 4372800. F: +971 (0)4 4372818. France: State          Unless otherwise stated, all data and all data sourced to “UN” is the most current
Street Global Advisors France. Authorised and regulated by the Autorité des Marchés      data attributable to the United Nations Population Division. Data shown beyond 2015
Financiers. Registered with the Register of Commerce and Companies of Nanterre           is a UN projection.
under the number 412 052 680. Registered office: Immeuble Défense Plaza, 23-25
                                                                                         This communication is directed at professional clients (this includes eligible
rue Delarivière-Lefoullon, 92064 Paris La Défense Cedex, France. T: (+33) 1 44 45 40
                                                                                         counterparties as defined by the appropriate EU regulator or applicable Swiss
00. F: (+33) 1 44 45 41 92. Germany: State Street Global Advisors GmbH, Brienner
                                                                                         regulator) who are deemed both knowledgeable and experienced in matters relating
Strasse 59, D-80333 Munich. Authorised and regulated by the Bundesanstalt für
                                                                                         to investments. The products and services to which this communication relates are
Finanzdienstleistungsaufsicht (“BaFin”). Registered with the Register of Commerce
                                                                                         only available to such persons and persons of any other description (including retail
Munich HRB 121381. T: +49 (0)89 55878 400. F: +49 (0)89 55878 440. Hong Kong:
                                                                                         clients) should not rely on this communication.
State Street Global Advisors Asia Limited, 68/F, Two International Finance Centre,
8 Finance Street, Central, Hong Kong. T: +852 2103 0288. F: +852 2103 0200.              Unless otherwise noted, the opinions expressed are those of the author who is a
Ireland: State Street Global Advisors Ireland Limited is regulated by the Central        researcher at State Street. Views and opinions are subject to change at any time
Bank of Ireland. Registered office address 78 Sir John Rogerson’s Quay, Dublin 2.        based on market and other conditions.
Registered number 145221. T: +353 (0)1 776 3000. F: +353 (0)1 776 3300. Italy:           This document contains certain statements that may be deemed forward-looking
State Street Global Advisors Limited, Milan Branch (Sede Secondaria di Milano) is        statements. Please note that any such statements are not guarantees of any
a branch of State Street Global Advisors Limited, a company registered in the UK,        future performance and actual results or developments may differ materially from
authorised and regulated by the Financial Conduct Authority (FCA ), with a capital       those projected.
of GBP 71’650’000.00, and whose registered office is at 20 Churchill Place, London       The information provided does not constitute investment advice and it should not be
E14 5HJ. State Street Global Advisors Limited, Milan Branch (Sede Secondaria di          relied on as such. You should consult your tax and financial advisor.
Milano), is registered in Italy with company number 06353340968 - R.E.A. 1887090
and VAT number 06353340968 and whose office is at Via dei Bossi, 4 - 20121 Milano,       All information has been obtained from sources believed to be reliable, but its
Italy. T: 39 02 32066 100. F: 39 02 32066 155. Japan: State Street Global Advisors       accuracy is not guaranteed. There is no representation or warranty as to the current
(Japan) Co., Ltd., Toranomon Hills Mori Tower 25F 1-23-1 Toranomon, Minato-ku,           accuracy, reliability or completeness of, nor liability for, decisions based on such
Tokyo 105-6325 Japan. T: +81 3 4530 7380. Financial Instruments Business Operator,       information and it should not be relied on as such.
Kanto Local Financial Bureau (Kinsho #345), Membership: Japan Investment Advisers        The whole or any part of this work may not be reproduced, copied or transmitted or
Association, The Investment Trust Association, Japan, Japan Securities Dealers’          any of its contents disclosed to third parties without SSGA’s express written consent.
Association. Netherlands: State Street Global Advisors Netherlands, Apollo

                                                                                                                       © 2018 State Street Corporation. All Rights Reserved.
                                                                                                                  ID12400-2020370.2.1.GBL.RTL 0318 Exp. Date: 31/03/2019

State Street Global Advisors                                                                                                                                                24
You can also read